Friday, August 19, 2011

Guest Post from Donovan Moore: Is Matt Wieters Tired at the Plate?

Another guest post form Donovan and I thank him. I am also supposed to pass along that the August stats referenced in this post are through August 11th only. Enjoy!

This kind of seems like a funny question… what is being asked? Do we catch glimpses of the young catcher yawning in between pitches? Is he taking afternoon naps in the clubhouse? While those are certainly hilarious images—this is not what I mean.

More and more frequently I have been hearing conjectures and comments about how Matt Wieters’s offensive performance “could” or “would” be “better” if he wasn’t so tired out from catching so many games. On its face, this argument seems like it has the potential to be true. A young guy in only his second full Major League season at the most demanding position on the diamond must need some sort of grace period to get used to the wear and tear that catching puts on his body. The expectations put on Wieters’s expected offensive performance when coming through the ranks in the Minors were so extremely out-of-this-world that for him to have fallen so short of them must demand an explanation of sorts…right?

I decided that it could not hurt to attempt to investigate this subject statistically. To begin, I had to determine which stats I would investigate in order to see if the numbers would lend themselves to the argument that Wieters is in fact tired at the plate. The offensive numbers that I decided to take a look at were batting average (AVG) and slugging percentage (SLG). Batting average is the most basic statistic kept on hitters—what percentage of at bats does this guy get a base knock? However, I needed a statistic to qualify these base knocks--and that’s where SLG comes in. SLG is a measurement of how many total bases a batter gets per plate appearance, on average. The differences between a single, a double, a triple, and a home run are all given equal weight. A guy could have an AVG of .350, but if he only hits singles, it’s highly unlikely that he will be nearly as valuable as a guy who is batting .250, but hits a home run every hit.

When looking at SLG and AVG, it made perfect sense to me that if Wieters was in fact tired, he would get less hits (AVG) and they wouldn’t go as far (SLG). Having established which measures of offensive production will be used is one thing, but how does one measure how tired any given player is on any given day? I devised two ways to attempt to measure Wieters’s level of “tired” in any given game. In order to do this, I categorized every game that Matt has played in the 2011 season as either, “he didn’t have the previous day off,” “he had the previous day off,” or “he had the previous day and at least one more consecutive day before the previous day off”—including the first game of the season and the game immediately following the AS break. I took SLG and AVG numbers that he posted in games when he had the previous day off and compared them to his season base line.

This is what I came up with:

2011 season base line: .256 AVG .391 SLG

2011 “he had the previous day off”: .254 AVG .377 SLG

2011 “he had the previous day+ off”: .277 AVG .556 SLG

The only number among these that look anomalous is the .556 SLG when Matt has had more than one day off in a row. However, the other fishy number that goes along with having more than one day off for Wieters is his BABIP (batting average on balls in play). In these games, Wieters has a .357 BABIP. Given his LD% (line drive rate) in these games, his expected BABIP is .334. The couple of extra hits on these days have simply been a couple of extra line drives falling in for doubles. Wieters’s season LD% is 18.2 (for an expected BABIP of .302). His LD% in these games is 21.4%. He is outperforming his expected BABIP by .023 on during these games. During the season as a whole he is underperforming his expected BABIP of .302 by -0.016 with a .286 mark. Given the fact that these games where he’s been sufficiently rested have only amounted to 14 at bats for Wieters, I am pretty comfortable saying that the spike in SLG is due in part to small sample size, and in part to the fact that his BABIP (largely a “luck” statistic) is just simply higher on these days. Simply stated: Wieters have just been luckier in these 14AB than he has been during the rest of the season.

So, Matt is not performing phenomenally better on days after he has been rested or had a break. The other way to measure whether he is tired at the plate or not is to compare his season statistics per month. If he is tiring out over time, we should see a decline here. This is what I came up with after doing so:

April:  .260AVG /.493 SLG

May:    .278AVG /.344 SLG

June:   .247AVG /.383 SLG

July:   .235AVG /.376 SLG

August: .262AVG /.357 SLG

…or graphically:

Wieters AVG has more or less been stable monthly. The only number that really sticks out here is his .493 SLG in April. One could argue that after slugging the crap out of the ball in the first month of the season, Matt just got tired. However, I feel pretty safe saying that if Wieters truly was tired that the drop off in SLG would be steady instead of being drastic one month and then rebounding slightly again. I’m not comfortable saying that one outlier has any statistical significance.

The conjecture that Wieters is tired at the plate makes perfect sense on its face. However the stats just simply do not lend themselves to this argument whatsoever. Who knows whether this level of production will persist throughout his career, or if there is a big break out on the horizon for the Orioles’ young catcher? Either way, if he can maintain his current level of defense, he really doesn’t need to hit very much at all to be a valuable player.

Tuesday, August 2, 2011

Guest Post at There's No Defense

While my posts here are few these days, I am still posting over on the Orioles Buzz blog at This last post was about the woeful Oriole defense. Enjoy?

Monday, August 1, 2011

Guest Post from Donovan Moore: Freaky Friday Swap - The Luck of the Starting Pitcher

Hello again. This is yet another guest post by Donovan Moore, who I again thank profusely for giving some of his effort to keep some new content on here until I get back on track. Donovan gave this to me on Friday evening but I was working all weekend and did not see it until this morning. So any outdated references are completely my fault. But the meat of this post remains relevant...and fun.

It’s officially that time of year in baseball when following the Baltimore Orioles has become something that legitimately just really kind of gets you down. Posts about how bad the Orioles are playing aren’t particularly interesting to write. Lately, that’s about all there has been to write about—and consequently that’s all there has been to read about. As someone who not only writes about the Orioles but also reads a great deal about them as well, I have decided that at least in the immediate time being I am going to try to shift focus in the blog-O-sphere away from a tone of sulking, whining, and second-guessing. I’m not starting any sort of campaign against negative writing, because really the bottom line is that when the team one writes about is a really bad team, often times a disheartened reader would like to read someone write the angry things that he is himself thinking. I am also not attempting to write with a tone of seeming like I think that I am better than anyone who is writing negative things about the Orioles. The team is bad…in pretty much almost every way. We get it. I am not refuting it.

In the next couple of days, the non-waiver trade deadline will come and go and either the Orioles will make a few swaps, or they won’t. After the Orioles either make a couple of deals—or not—there will be a flurry of posts about the trades they managed to pull off, the treades they didn’t manage to pull off, and the trades that they didn’t manage to even look into (Colby Rasmus?). It’s really sort of difficult to not get sucked into the same day-to-day moaning and groaning that seems to come along with covering this Orioles team, however in order to do so, let’s imagine an alternate reality (you’ve got my attention, now…)!

This alternate reality is not one where the Orioles are actually a very good team, because—well I am not trying to hurt anyone’s brain just quite yet. This alternate reality is one where mainstream baseball recognizes the complete uselessness of the W/L statistic for pitchers! This idea came to mind when, not just but a few days ago, the on-air announcers for the Orioles game mentioned that the Orioles have scored nine runs or more six times this year, and that Arrieta had been the starting pitcher four of those times. This game became the seventh time the Orioles scored nine or more runs—and Arrieta was the starting pitcher again.

I would assume that any person reading this blog is already of the opinion that the W/L statistic for pitchers (particularly starting pitchers) is almost completely useless. The point of this post is not to argue this point. The point of this post is to demonstrate through example that the W/L statistic is rooted pretty deeply in luck. How will this be accomplished? Well, I am going to pull off a Freaky-Friday-type swap of the luck of Jeremy Guthrie and the luck of Jake Arrieta. Again, I am not formulating this as any sort of logical argument for or against the W/L statistic—I am merely using my magical luck-swapping skills in a demonstration purely for the enjoyment and excitement of onlookers. Let me also say that my intention is not to pick on Jake Arrieta. I chose these two pitchers for reasons that are pretty obvious. Jeremy Guthrie in this season (and really in other past seasons, too) has suffered from poor run support and terrible bullpen intervention. Despite the fact that he is pitching to a very respectable 4.33ERA, 4.36FIP, and 4.31xFIP he is leading THE ENTIRE MLB in losses (4-14). This really just does not seem fair at all. The won/loss record of a starting pitcher is typically the first one any person hears that is associated with the given pitcher. Guthrie’s 4-14 is simply not a good reflection at all of how he has pitched in 2011. On the other hand, there is Jake Arrieta. Arrieta is not the most lucky pitcher in the MLB or AL, but he does in fact receive the highest run support of any starting pitcher on the 2011 Orioles (Guthrie gets the lowest). For this reason, he is the obvious choice to be Guthrie’s counterpart in this Freaky-Friday-type swap. Arrieta’s current won/loss record is 10-7. He boasts a 5.12ERA, 5.28FIP, and a 4.31xFIP. These numbers are not awful by any measure, but they are worse than Guthrie’s—even though he mysteriously has a much healthier looking 10-7 won/loss record. Additionally (as of 7/28), each pitcher has made 21 starts on the year, with 11 of them being “quality” starts (6IP or more with 3 or less ER). Quality starts are a pretty good measure of what sort of chance a starting pitcher has given his team to win—with obvious exceptions, such as pitching 5 perfect innings only to get injured and leave the game.

Okay…time to wave the magic wand.

Bizarro Guthrie: 12-8 W/L record
Bizarro Arrieta: 5-12 W/L record

In order to achieve these figures, all I did was match up Guthrie and Arrieta’s starts (numbered 1-21) to each corresponding start and swap out their pitching performances in order to come up with a W, L, or nd. The bullpen performances after each of the pitchers left the game was left the same, as were the offensive performances of the Orioles.

This is quite obviously not a perfect measure because it fails to take into account many important factors that are impossible to adjust for. For instance, it is not exactly fair to correspond Arrieta’s performance against Seattle with Guthrie’s performance against Boston—but this is the best we’ve got without any actual magic.

A couple interesting things that I came across while plowing through all of the numbers for each of the 21 games that these two men have started in 2011:

Arrieta has only pitched into the 7th inning four times this year. He has not pitched past the 7th once.

Guthrie has one win stemming from a non-quality start. Arrieta has three.

Guthrie has eight (EIGHT!?!?) losses or non-decisions in a quality start. Arrieta has only one.

Due to poor run-support and bad performances by the bullpen, Jeremy Guthrie (the better pitcher of the two) has a far poorer W/L record than Jake Arrieta. 4-14 and 10-7 really do appear to be worlds apart—and neither truly reflects the true performances of the pitcher that they are assigned to. Arrieta has been luckier than your average bear, while Guthrie has been unluckier. If there were somehow a way to make W/L a fair measure of the performance of any given starting pitcher, each of these gentlemen would have their mark sitting somewhere between what it is now, and what their bizarro counterpart’s mark is.

Unfortunately, there is not really a way to make this statistic fair—it relies too heavily on things that are out of the control of the starting pitcher. Guthrie’s 14 losses lead all of the Majors. Given Arrieta’s luck, Guthrie’s 12 wins would be third best in the AL.